48415 20200221144151.0 doi 10.1016/j.imavis.2015.12.006 sideral 94226 ART-2016-94226 eng (orcid)0000-0001-6324-940X Rodriguez, Mario A Time-Flexible-Kernel framework for video-based activity recognition 2016 Access copy available to the general public Unrestricted This work deals with the challenging task of activity recognition in unconstrained videos. Standard methods are based on video encoding of low-level features using Fisher Vectors or Bag of Features. However, these approaches model every sequence into a single vector with fixed dimensionality that lacks any long-term temporal information, which may be important for recognition, especially of complex activities. This work proposes a novel framework with two main technical novelties: First, a video encoding method that maintains the temporal structure of sequences and second a Time Flexible Kernel that allows comparison of sequences of different lengths and random alignment. Results on challenging benchmarks and comparison to previous work demonstrate the applicability and value of our framework. info:eu-repo/grantAgreement/ES/MINECO/TIN2013-45312-R info:eu-repo/semantics/openAccess by-nc-nd http://creativecommons.org/licenses/by-nc-nd/3.0/es/ 2.671 2016 COMPUTER SCIENCE, THEORY & METHODS 19 / 104 = 0.183 2016 Q1 T1 COMPUTER SCIENCE, SOFTWARE ENGINEERING 17 / 106 = 0.16 2016 Q1 T1 OPTICS 26 / 92 = 0.283 2016 Q2 T1 ENGINEERING, ELECTRICAL & ELECTRONIC 75 / 260 = 0.288 2016 Q2 T1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE 37 / 133 = 0.278 2016 Q2 T1 0.953 2016 Computer Vision and Pattern Recognition 2016 Q1 Signal Processing 2016 Q1 Electrical and Electronic Engineering 2016 Q1 info:eu-repo/semantics/article info:eu-repo/semantics/acceptedVersion (orcid)0000-0002-0903-5520 Orrite, Carlos Universidad de Zaragoza (orcid)0000-0001-7671-7540 Medrano, Carlos Universidad de Zaragoza Makris, Dimitrios 5008 785 Universidad de Zaragoza Dpto. Ingeniería Electrón.Com. Área Tecnología Electrónica 48-49 (2016), 26-36 Image vis. comput. IMAGE AND VISION COMPUTING 0262-8856 7686556 http://zaguan.unizar.es/record/48415/files/texto_completo.pdf Postprint 48976 http://zaguan.unizar.es/record/48415/files/texto_completo.jpg?subformat=icon icon Postprint oai:zaguan.unizar.es:48415 articulos driver 2020-02-21-13:07:37 ARTICLE